Logically right, practically wrong: how Kana Rain came to be

How Kana Rain came to be

Dating a nisei woman from a traditional Japanese family is, without a doubt, a blessing. But it comes with its challenges, one of them being having to learn Japanese to communicate with her family.

Languages are cool, but they're not exactly where my hyperfocus tends to go. But I like playing games. And I like creating things. So I decided to build what would become the first in a series of games: Kana Rain.

Kana Rain is a very simple game designed to help you learn and memorize the shape, sound, and romaji transcription of each hiragana and katakana character (I'll just call them kana), which make up a small but fundamental part of the Japanese writing system.

To better understand how it works, check out the game's README on GitHub.

In the game, raindrops containing kana fall from the sky, and the player has to type their equivalent in the Roman alphabet to collect them and gradually expand their collection of known characters (which the game calls a garden).

With three-minute sessions to keep things from getting boring or exhausting, and three lives per round, the player is progressively exposed to each kana. If you play for about 30 minutes a day, you can learn all of them in roughly two weeks. That's exactly how I learned them.

If you'd like to play and learn too, you can try it directly in your browser, or watch a short demonstration on X.

AI skepticism almost killed my game

The fact is, the game works. You can easily run it locally if you want to, and it's fun (I know, I'm biased).

However, before I started working on it, I did some research with AI about the world of educational games. And the AIs made a pretty strong case that what I was planning probably wasn't a good idea, because:

1. memorizing kana inside the game wouldn't necessarily translate into remembering and recognizing them outside the game;
2. this type of game wouldn't contribute much to some esoteric set of metrics for Japanese language learning.

I considered giving up because I didn't want to put effort into something that would lead nowhere.

How AI unlocks creative potential

But then I remembered that trying would cost me very little.

Two years ago, building a game like this would have been a painful, multi-day undertaking. But with AI by my side, the core of the game could be built in just a few hours. In the end, it took me a few hours initially, plus a few more over the following days, mostly making adjustments until the gameplay felt right.

Since the cost of building it was so low, I went ahead and did it. And I had fun building it, then playing it. And I came out the other side having fully learned all the Japanese kana.

This project was never meant to make money. It was simply meant to help me learn. I learned this small part of Japanese and, as a bonus, an important lesson that I bet will make a difference in the years to come.

The skeptic is usually right, but the optimist is usually successful.

I once heard that saying, and it has been stuck in my head for a long time. Little by little, it keeps coming back and making a bit more sense.

The AIs were right that my game's methodology wasn't academically the best. But at the same time, I was wrong to think, even for a second, that an apparently flawed idea, one that my gut was urging me to pursue, wasn't worth trying.

Looking back at my life so far, I believe I've been very conservative, very risk-averse. Over the past two years, though, I've gone through a series of experiences that opened my mind to the possibility of taking more risks. And this latest episode, an experiment with a simple educational game, has inspired me to crystallize that realization in this blog post.

If I hadn't built the game because I thought it wouldn't lead anywhere, I would have been logically right, but practically wrong. I would have been right that Kana Rain isn't some groundbreaking educational game, but wrong not to have put in a little sweat to bring it to life.

Because that sweat gave meaning to the game's inherent flaws and gave me something beyond technical knowledge: it gave me wisdom as a creator.